Describes the design and implementation of a large-scale experimental methodology for testing the linguistic knowledge of Persian heritage language learners.

Presents the design of a comprehensive test battery for assessing Persian linguistic competence across phonology, morphology, syntax, and broader communicative abilities. Describes elicitation tasks, sample items, and experimental controls for investigating learners’ knowledge and the influence of English.

Evaluation of real-time, two-way English–Iraqi Arabic and English-Dari speech translation systems under DARPA’s TRANSTAC program. Testing in laboratory and outdoor scenarios assessed speech recognition, translation, noise robustness, usability, and performance on portable hardware. (not public released)

Examines whether Persian noun–verb combinations require noun incorporation or pseudo-incorporation. Argues that the relevant properties of nonspecific nominal arguments follow from their internal structure and position within the verbal domain, without a separate incorporation mechanism.

Overview of the July 2007 TRANSTAC evaluation of English–Iraqi Arabic and English-Dari speech translation technology. Focuses on assessing prototype systems for spoken communication in tactical settings, using controlled laboratory tests and outdoor scenarios with realistic background noise. (not public released)

Contrasts morphological and analytic causatives in Eastern Armenian using evidence from binding, adverbial scope, and causee behavior. Proposes a single-event structure for morphological causatives and a two-event structure for analytic constructions.

Examines Eastern Armenian voice morphology, focusing on the suffix shared by reflexive, passive, and anticausative constructions. Investigates how the same morphological form relates to constructions with distinct syntactic and semantic properties.

Develops a preliminary typology of Persian complex predicates based on component contributions, transitivity, and aspect. Examines several light verbs and preverbal element types, proposing structural analyses of causation, activity, instrumental relations, and event delimitation.

Reviews heritage language research relevant to designing listening, speaking, and literacy assessments, with an emphasis on Persian. Also considers instruments for measuring the sociolinguistic factors that influence heritage language learning.

Proposes semantic templates for representing Persian light verb constructions in multilingual NLP. Models primitive meaning components and argument structure to handle expressions whose parts can be separated in syntax, with templates serving as interlingual representations.

Presents a two-level Persian morphological analyzer built with Xerox finite-state tools. Addresses complex verbal conjugation, long-distance morphological dependencies, phonological alternations, and ambiguous word boundaries caused by variable spacing.

Introduces corpus-based approaches in computational linguistics for a linguist audience.